CN107808381A - A kind of unicellular image partition method - Google Patents
A kind of unicellular image partition method Download PDFInfo
- Publication number
- CN107808381A CN107808381A CN201710871540.0A CN201710871540A CN107808381A CN 107808381 A CN107808381 A CN 107808381A CN 201710871540 A CN201710871540 A CN 201710871540A CN 107808381 A CN107808381 A CN 107808381A
- Authority
- CN
- China
- Prior art keywords
- segmentation
- image
- mrow
- threshold
- result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20004—Adaptive image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention discloses a kind of unicellular image partition method, and this method comprises the following steps:1) image preprocessing, gray level image is converted into, removes noise and carry out contrast enhancing;2) carry out piecemeal Threshold segmentation and segment the image into A*A part fritters, every piece calculates optimal threshold using OSTU, is partitioned into prospect and background;3) judge by the cell nuclear state that upper step is split to obtain be characterized in it is no meet normal, prove that the segmentation result is that comparison is good if normal, output result;4) if segmentation result does not meet cell nuclear state feature, result segmentation is not accurate, carries out next step image procossing;5) adaptive threshold fuzziness is carried out, and the output result together with other normal segmentation figure pictures, a kind of unicellular partitioning algorithm of the present invention, it is inaccurate to solve nucleus segmentation, the slow problem of splitting speed, the characteristics of piecemeal Threshold segmentation speed is fast and adaptive threshold fuzziness precision is high, workload is few is combined, both form mutual supplement with each other's advantages, improve the quality of output image.
Description
Technical field
The present invention relates to Methods of Segmentation On Cell Images and identification technology field, specially a kind of unicellular image partition method.
Background technology
Using image processing techniques and the clinical experience of pathology expert, cervical cancer cell image can be carried out quick
Examination and statistics, and then the automatic diagnosis to cervical cell image is realized, the efficiency of doctor can be thus greatly improved, reduces people
The error and erroneous judgement of work diagosis, at present to cervical cell image mainly to realize based on automatic segmentation and classification, according to cell kind
Class employs various kinds of cell image segmentation algorithm, and wherein Threshold segmentation is most widely used and most simple among image segmentation
A kind of dividing method.
It is not preferable thin but because the subjective factor of film-making influences, most of cervical cell image is all more complicated
The such background of born of the same parents' image is single and does not have impurity, each cell be individually present and easily it is discernable, this causes subsequently
Image segmentation and cell classification bring many difficulties.Major Difficulties have:1) iuntercellular is overlapping;2) cell edges boundary obscures;
3) cell image participates in impurity;4) nucleus size shape texture is inconsistent.
Cell segmentation mainly has piecemeal Threshold segmentation and adaptive threshold fuzziness, because its algorithm arithmetic speed is fast and splits
The features such as accurate, it is set to obtain the favor of researcher in image segmentation field.Currently in cell image segmentation method, according to
Cellular morphology has been widely used by the way of many algorithms are combined.
The content of the invention
It is an object of the invention to provide a kind of unicellular image partition method, to solve what is proposed in above-mentioned background technology
Problem.
To achieve the above object, the present invention provides following technical scheme:A kind of unicellular image partition method, including it is following
Step:
Partitioning algorithm overall framework of the present invention is introduced first, is split first with improved piecemeal threshold value
Algorithm obtains relatively rough segmentation result, and then using nuclear characteristics form method of testing, piecemeal threshold value is split
The result obtained afterwards is screened, and incomplete nucleus is sent into adaptivenon-uniform sampling and carries out secondary splitting, complete nucleus figure
As directly exporting, then the result of the result using adaptivenon-uniform sampling and piecemeal Threshold segmentation is merged into final preferable output
Image.
Then, then improvement Threshold Segmentation Algorithm step of the present invention is introduced, improvement threshold value of the present invention is divided
Cut algorithm and include piecemeal Threshold segmentation and adaptive threshold fuzziness two parts, the basic thought of piecemeal Threshold Segmentation Algorithm is thin
Born of the same parents' image is divided into some equal fritters, and global threshold dividing method threshold value is utilized to each small piecemeal, is partitioned into every
The background and prospect of one fritter, it is exactly dividing for whole cell image that finally the segmentation figure picture of some small piecemeals, which is merged together,
Result is cut, and the basic thought of adaptivenon-uniform sampling algorithm is that Threshold segmentation is all set to all pixels point in image, each picture
For the threshold value of vegetarian refreshments all centered on the pixel, neighborhood window is the segmentation threshold that size calculates the window, and specific formula is as follows:
T (x, y)=Mean=∑s(x,y)∈wI(x,y)|w (2)
Brief description of the drawings
Fig. 1 is dividing method overall framework flow chart;
Fig. 2 is improved Threshold Segmentation Algorithm flow chart;
Embodiment
It will start to combine the accompanying drawing in the embodiment of the present invention below, the technical scheme in the embodiment of the present invention carried out
It is whole, clearly state.Obviously, the example stated is only part of the embodiment of the present invention, rather than whole embodiments;Base
Embodiment in the present invention, what those of ordinary skill in the art were obtained on the premise of creative work is not made
All other examples, belong to protection scope of the present invention.
Referring to Fig. 1, the present invention provides a kind of technical scheme:A kind of unicellular image partition method, including overall segmentation
Method and improve cell threshold value partitioning algorithm, main flow for input one width uterine neck TCT images, to image carry out gradation conversion,
Then all images are carried out piecemeal Threshold segmentation processing, used by the pretreatment such as medium filtering denoising and contrast enhancing
OTSU methods enter row threshold division, are partitioned into prospect and background, then carry out picture shape test, contexts image distinctness
Result output set to be detected is directly placed into, not by the carry out adaptive threshold fuzziness again of detection, then splitting twice
Result merge, just obtain final preferable segmentation image collection, the present invention as a result of two kinds of dividing methods, according to
Cellular morphology local feature uses different partitioning algorithms, can become apparent from the karyomorphism after segmentation, accurately, reduces
The quantity of cell processing, the speed of algorithm is improved, cell pathology analysis is carried out and classification is provided convenience to be follow-up, wherein
Piecemeal Threshold segmentation has main steps that:With reference to figure 2, fritter one by one is divided the image into, gray scale is carried out respectively to each piece
Change, calculate variance V and average Mean, the optimal threshold T1 and T2 of entire image are obtained using big rule algorithm, if variance V is more than
T1, then global threshold segmentation is done using OTSU, if variance V is less than T1, and average Mean is more than T2, then can determine whether the fritter category
In background area, otherwise just belong to foreground area, and adaptivenon-uniform sampling mainly comprises the following steps:Gray scale is carried out to image first
Processing, then travels through each pixel I (x, y), and segmentation threshold T is set to each pixel, using the pixel as core,
Neighborhood window is set, obtains the gray average Mean of the window, the background of the image is then obtained with before according to contrast can
Scape.Specific formula is as follows:
T (x, y)=Mean=∑s(x,y)∈wI(x,y)|w (2)
It should be apparent to those skilled in the art that the invention is not restricted to the details of above-mentioned one exemplary embodiment, and do not taking off
On the premise of spirit or essential attributes from the present invention, it can realize that all changes of the present invention include with other concrete modes
In the present invention, any reference in claim should not be seen as and limit related claim invention, therefore,
No matter for which aspect, embodiment should be all considered as it is exemplary, and be it is nonrestrictive, the scope of the present invention by
Appended claims rather than described above limit, it is intended that will fall scope and implication in the equivalency of claim
In.
It is not every kind of embodiment in addition, reason is it should be appreciated that although this specification is stated according to embodiment
An independent technical scheme is only included, this narrating mode of specification is only those skilled in the art for clarity
Should be using specification as an entirety, the technical scheme in each embodiment can also pass through appropriate combination, form this area
The other embodiment that technical staff is appreciated that.
Claims (3)
1. a kind of unicellular image partition method, including overall segmentation method and improvement cell threshold value partitioning algorithm, main flow:
1) a width uterine neck TCT images are inputted, the pretreatments such as gradation conversion, medium filtering denoising and contrast enhancing are carried out to image;
2) piecemeal Threshold segmentation processing is carried out to all images, row threshold division is entered using OTSU methods, is partitioned into prospect and background;3)
Picture shape test is carried out, contexts image distinctness is directly placed into result output set to be detected, not by detection again
Secondary carry out adaptive threshold fuzziness;4) result split twice is merged, just obtains final preferably segmentation image collection,
This programme carries out the segmentation of distinct methods according to cell different shape respectively, ensure that the accuracy and independence of dividing method,
Wherein piecemeal Threshold segmentation has main steps that:1) divide the image into fritter one by one, to each piece respectively carry out gray processing,
Calculate variance V and average Mean;2) the optimal threshold T1 and T2 of entire image are obtained using big rule algorithm;3) if variance V is big
In T1, then global threshold segmentation is done using OTSU, if variance V is less than T1, and average Mean is more than T2, then can determine whether the fritter
Belong to background area, otherwise just belong to foreground area, and adaptivenon-uniform sampling mainly comprises the following steps:1) image is carried out first
Gray proces;2) each pixel I (x, y) is traveled through, segmentation threshold T is set to each pixel, using the pixel as core
The heart, neighborhood window is set, obtains the gray average Mean of the window;3) background of the image is obtained with before according to contrast can
Scape;Specific formula is as follows:
<mrow>
<mi>G</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mn>0</mn>
<mo>,</mo>
<mi>I</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<mo>=</mo>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mn>1</mn>
<mo>,</mo>
<mi>I</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>></mo>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
T (x, y)=Mean=∑s(x,y)∈wI(x,y)|w (2)
A kind of 2. unicellular image partition method according to claim 1, it is characterised in that:The Methods of Segmentation On Cell Images side
Method uses different partitioning algorithms, after can making segmentation as a result of two kinds of dividing methods according to cellular morphology local feature
Nucleus become apparent from and accurately, reduce the quantity of cell processing, improve the speed of algorithm, cytopathy is carried out to be follow-up
Reason analysis and classification are provided convenience.
A kind of 3. unicellular image partition method according to claim 1, it is characterised in that:The improved Threshold segmentation
Algorithm fully according to the actual local feature of cell, combines fast piecemeal Threshold segmentation speed, segmentation image clearly and adaptive thresholding
It is worth the characteristics of segmentation precision is high, workload is few, both form mutual supplement with each other's advantages, improve the quality of output image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710871540.0A CN107808381A (en) | 2017-09-25 | 2017-09-25 | A kind of unicellular image partition method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710871540.0A CN107808381A (en) | 2017-09-25 | 2017-09-25 | A kind of unicellular image partition method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107808381A true CN107808381A (en) | 2018-03-16 |
Family
ID=61584519
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710871540.0A Pending CN107808381A (en) | 2017-09-25 | 2017-09-25 | A kind of unicellular image partition method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107808381A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108596932A (en) * | 2018-04-18 | 2018-09-28 | 哈尔滨理工大学 | A kind of overlapping cervical cell image partition method |
CN108921868A (en) * | 2018-07-02 | 2018-11-30 | 中央民族大学 | A kind of improved Otsu threshold segmentation method |
CN109064475A (en) * | 2018-09-11 | 2018-12-21 | 深圳辉煌耀强科技有限公司 | For the image partition method and device of cervical exfoliated cell image |
CN109191470A (en) * | 2018-08-18 | 2019-01-11 | 北京洛必达科技有限公司 | Image partition method and device suitable for big data image |
CN109191434A (en) * | 2018-08-13 | 2019-01-11 | 阜阳师范学院 | Image detecting system and detection method in a kind of cell differentiation |
CN109493330A (en) * | 2018-11-06 | 2019-03-19 | 电子科技大学 | A kind of nucleus example dividing method based on multi-task learning |
CN109801303A (en) * | 2018-12-18 | 2019-05-24 | 北京羽医甘蓝信息技术有限公司 | Divide the method and apparatus of cell in hydrothorax fluorescent image |
CN110087063A (en) * | 2019-04-24 | 2019-08-02 | 昆山丘钛微电子科技有限公司 | A kind of image processing method, device and electronic equipment |
CN110111307A (en) * | 2019-04-12 | 2019-08-09 | 张晓红 | A kind of immune teaching immune system feedback analog system and method |
CN110570445A (en) * | 2019-09-09 | 2019-12-13 | 上海联影医疗科技有限公司 | Image segmentation method, device, terminal and readable medium |
CN112270370A (en) * | 2020-11-06 | 2021-01-26 | 北京环境特性研究所 | Vehicle apparent damage assessment method |
CN117252893A (en) * | 2023-11-17 | 2023-12-19 | 科普云医疗软件(深圳)有限公司 | Segmentation processing method for breast cancer pathological image |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102682305A (en) * | 2012-04-25 | 2012-09-19 | 深圳市迈科龙医疗设备有限公司 | Automatic screening system and automatic screening method using thin-prep cytology test |
CN102831607A (en) * | 2012-08-08 | 2012-12-19 | 深圳市迈科龙生物技术有限公司 | Method for segmenting cervix uteri liquid base cell image |
CN103489187A (en) * | 2013-09-23 | 2014-01-01 | 华南理工大学 | Quality test based segmenting method of cell nucleuses in cervical LCT image |
CN103984958A (en) * | 2014-05-07 | 2014-08-13 | 深圳大学 | Method and system for segmenting cervical caner cells |
CN104992435A (en) * | 2015-06-24 | 2015-10-21 | 广西师范大学 | Cervix uteri single cell image segmentation algorithm |
US20170193657A1 (en) * | 2015-12-30 | 2017-07-06 | Case Western Reserve University | Prediction of recurrence of non-small cell lung cancer with tumor infiltrating lymphocyte (til) graphs |
-
2017
- 2017-09-25 CN CN201710871540.0A patent/CN107808381A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102682305A (en) * | 2012-04-25 | 2012-09-19 | 深圳市迈科龙医疗设备有限公司 | Automatic screening system and automatic screening method using thin-prep cytology test |
CN102831607A (en) * | 2012-08-08 | 2012-12-19 | 深圳市迈科龙生物技术有限公司 | Method for segmenting cervix uteri liquid base cell image |
CN103489187A (en) * | 2013-09-23 | 2014-01-01 | 华南理工大学 | Quality test based segmenting method of cell nucleuses in cervical LCT image |
CN103984958A (en) * | 2014-05-07 | 2014-08-13 | 深圳大学 | Method and system for segmenting cervical caner cells |
CN104992435A (en) * | 2015-06-24 | 2015-10-21 | 广西师范大学 | Cervix uteri single cell image segmentation algorithm |
US20170193657A1 (en) * | 2015-12-30 | 2017-07-06 | Case Western Reserve University | Prediction of recurrence of non-small cell lung cancer with tumor infiltrating lymphocyte (til) graphs |
Non-Patent Citations (1)
Title |
---|
何小臻: "基于质量检测的宫颈细胞学图像的细胞核分割方法", 《中国优秀硕士学位论文全文数据库医药卫生科技辑》 * |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108596932A (en) * | 2018-04-18 | 2018-09-28 | 哈尔滨理工大学 | A kind of overlapping cervical cell image partition method |
CN108921868A (en) * | 2018-07-02 | 2018-11-30 | 中央民族大学 | A kind of improved Otsu threshold segmentation method |
CN108921868B (en) * | 2018-07-02 | 2021-08-24 | 中央民族大学 | Improved Otsu threshold segmentation method |
CN109191434A (en) * | 2018-08-13 | 2019-01-11 | 阜阳师范学院 | Image detecting system and detection method in a kind of cell differentiation |
CN109191470A (en) * | 2018-08-18 | 2019-01-11 | 北京洛必达科技有限公司 | Image partition method and device suitable for big data image |
CN109064475A (en) * | 2018-09-11 | 2018-12-21 | 深圳辉煌耀强科技有限公司 | For the image partition method and device of cervical exfoliated cell image |
CN109493330A (en) * | 2018-11-06 | 2019-03-19 | 电子科技大学 | A kind of nucleus example dividing method based on multi-task learning |
CN109801303A (en) * | 2018-12-18 | 2019-05-24 | 北京羽医甘蓝信息技术有限公司 | Divide the method and apparatus of cell in hydrothorax fluorescent image |
CN110111307A (en) * | 2019-04-12 | 2019-08-09 | 张晓红 | A kind of immune teaching immune system feedback analog system and method |
CN110111307B (en) * | 2019-04-12 | 2023-11-17 | 张晓红 | Immune system feedback simulation system and method for immune teaching |
CN110087063A (en) * | 2019-04-24 | 2019-08-02 | 昆山丘钛微电子科技有限公司 | A kind of image processing method, device and electronic equipment |
CN110570445A (en) * | 2019-09-09 | 2019-12-13 | 上海联影医疗科技有限公司 | Image segmentation method, device, terminal and readable medium |
CN110570445B (en) * | 2019-09-09 | 2022-03-25 | 上海联影医疗科技股份有限公司 | Image segmentation method, device, terminal and readable medium |
CN112270370A (en) * | 2020-11-06 | 2021-01-26 | 北京环境特性研究所 | Vehicle apparent damage assessment method |
CN112270370B (en) * | 2020-11-06 | 2023-06-02 | 北京环境特性研究所 | Vehicle apparent damage assessment method |
CN117252893A (en) * | 2023-11-17 | 2023-12-19 | 科普云医疗软件(深圳)有限公司 | Segmentation processing method for breast cancer pathological image |
CN117252893B (en) * | 2023-11-17 | 2024-02-23 | 科普云医疗软件(深圳)有限公司 | Segmentation processing method for breast cancer pathological image |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107808381A (en) | A kind of unicellular image partition method | |
CN112288706B (en) | Automatic chromosome karyotype analysis and abnormality detection method | |
CN108550133B (en) | Cancer cell detection method based on fast R-CNN | |
CN110428432B (en) | Deep neural network algorithm for automatically segmenting colon gland image | |
CN109523520B (en) | Chromosome automatic counting method based on deep learning | |
JP2021504816A (en) | Bone age evaluation and height prediction model, its system and its prediction method | |
CN110503630A (en) | A kind of cerebral hemorrhage classification, positioning and prediction technique based on three dimensional depth learning model | |
CN106056118A (en) | Recognition and counting method for cells | |
CN107610129B (en) | CNN-based multi-modal nasopharyngeal tumor joint segmentation method | |
CN104217213B (en) | A kind of medical image multistage sorting technique based on symmetric theory | |
CN110458835A (en) | A kind of image processing method, device, equipment and medium | |
CN108305253A (en) | A kind of pathology full slice diagnostic method based on more multiplying power deep learnings | |
TW202004776A (en) | Establishing method of bone age assessment and height prediction model, bone age assessment and height prediction system, and bone age assessment and height prediction method | |
CN103489187A (en) | Quality test based segmenting method of cell nucleuses in cervical LCT image | |
CN111429448A (en) | Bioluminescence target counting method based on weak segmentation information | |
CN112102250B (en) | Method for establishing and detecting pathological image detection model with training data as missing label | |
CN102855485A (en) | Automatic wheat earing detection method | |
CN108416304B (en) | Three-classification face detection method using context information | |
CN113139977A (en) | Mouth cavity curve image wisdom tooth segmentation method based on YOLO and U-Net | |
CN114693693A (en) | Fetal corpus callosum ultrasonic image segmentation method based on convolutional neural network | |
CN110378333B (en) | SD-OCT image macular fovea center positioning method | |
CN117292217A (en) | Skin typing data augmentation method and system based on countermeasure generation network | |
CN110889418A (en) | Gas contour identification method | |
CN113269747B (en) | Pathological image liver cancer diffusion detection method and system based on deep learning | |
CN114742803A (en) | Platelet aggregation detection method combining deep learning and digital image processing algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180316 |
|
WD01 | Invention patent application deemed withdrawn after publication |